#qi-hardware IRC log for Wednesday, 2012-03-28

qi-bot[commit] David Kühling: brainless: fix braindead comment copy-pasted from some other ackage (master) http://qi-hw.com/p/openwrt-packages/6ca100f01:24
qi-bot[commit] David Kühling: new package: alpy: python bindings for allegro game library (master) http://qi-hw.com/p/openwrt-packages/b2d16d901:24
qi-bot[commit] Adam Wang: changed AD1580BRTZ pin3 to Unspecified, added WM9707SCFT/V description with 48-lead TQFP, added few lib into .pro file (master) http://qi-hw.com/p/kicad-libs/c35e54d02:52
qi-bot[commit] Xiangfu: m1/patches/rtems: milkymist-audio-add-support-mic-boost.patch preserve the mic boost setting (master) http://qi-hw.com/p/wernermisc/56adfde03:24
qi-bot[commit] Adam Wang: added few libraries (master) http://qi-hw.com/p/kicad-libs/266c3b007:50
qi-bot[commit] Xiangfu: nanonotefiles: config.full_system update: include bard, gtypist remove tangogps, nanoterm, gtkperf (master) http://qi-hw.com/p/openwrt-packages/c102d4108:54
qi-botThe build has FAILED: http://fidelio.qi-hardware.com/~xiangfu/building/Nanonote/Ben/openwrt-xburst.minimal-20120327-0631 10:47
qi-bot[commit] Xiangfu: octave: disable curl (master) http://qi-hw.com/p/openwrt-packages/9c9945c11:59
dsafdsafanyone here?13:59
kristianpaulbard is cool14:01
kristianpauli noticed i allow me to resume my last reading where i was, thats fantastic !14:02
GNUtookristianpaul, what's bard?19:06
GNUtoohi btw19:06
GNUtooa nanonote ebook reader?19:06
qi-bot[commit] kyak: dega: fix build (master) http://qi-hw.com/p/openwrt-packages/4206f2919:09
qi-bot[commit] kyak: dgclock: fix build (master) http://qi-hw.com/p/openwrt-packages/e09fbf719:21
GNUtooduck duck go tells it's a package from openwrt19:34
kristianpaulGNUtoo: yup it is22:35
wolfspraulwpwrak: did you see Dick on the kicad list explaining that the company he works for (I guess) specifically says they don't want him to fix bugs or provide support anymore, only *new* features in KiCad? :-)23:25
wolfspraulthat is so funny23:25
wolfspraulthe often seen industry race - who can claim whatever feature X first, whether it works or not, whatever the details - all irrelevant23:26
wolfspraulby the time people find out feature X is totally broken, feature Y is being hyped and driving new sales :-)23:26
wpwrakah, interesting ...23:27
wpwrakfunny comparison with linus/linux. i'd say linux was already rather big by the time he started to earn real money with it. i guess C-based projects just scale more gracefully :)23:28
wolfspraulhe seems worn out23:28
wolfspraulshould sleep a bit :-)23:28
wolfspraulthe strategic concern of kicad should be readability and maintainability of code, modularity, community policies and culture, collaboration with geda, much better way for people to work on a high quality free component library, and a thousand other details23:29
wolfspraulmy 2c23:29
wolfspraulwhat Dick is describing there is mostly an illusion23:30
wolfspraulbut it's good that he's brining it up, so he shares his frustration a little and others can pad on his back23:30
wpwrakhis approach doesn't scale. it's the "core team of tired old men" approach.23:34
wolfspraulsure. also lack of trust in true community momentum.23:34
wolfspraulI think it was a bad day with his company telling him that he's not allowed to fix bugs, only add new features.23:34
wpwraka bad combination. a "core team" per se isn't so bad, but it has to be agile. else, it's only busy making excuses or fighting competition23:35
wpwrakwhile everybody else either gets frustated or falls asleep23:35
wolfspraulhonestly, with all my industry experience, that's one of the most stupidest things I've read in a long time, if that was really the mandate23:35
wolfspraulI cannot imagine any company on a sustainable path with that kind of mandate. I would go to the highest management available to me and put my badge on the table :-)23:36
wolfspraulbut I also think he is just describing it that way, his superiors were probably getting impatient about a lot of 'big features' that they want to see soon23:36
wolfspraulso yeah, a little stressed out there :-)23:37
wpwrakhmm, could be23:37
wpwrakwell, as long as the "core refactoring" is part of the "new features", a bit more focus wouldn't be so bad :)23:37
wolfspraulyou know when you hear about 'big refactoring', in >50% of cases that's the beginning of some trouble23:39
wolfspraulthe longer the refactoring takes the bigger the risk23:39
wpwrakyes, that refactoring has been going on for at least a year too long23:40
wpwraknothing wrong with discussing such things for a few days and then getting busy for some weeks. but at some point of time, the momentum just disappears23:41
wolfspraula year? then it's dead23:41
wolfspraulis the refactoring happening in live code/commits?23:41
wolfspraulI didn't notice anything23:41
wolfsprauloh, maybe I did notice the insane amount of renames back and forth23:42
wolfspraulwithout any consistency23:42
wolfspraulSomeFunction -> Some_Function -> some_function -> SomeFunc -> smFnc -> some_FUNCTION -> and so on23:42
wolfspraulthat's kicad23:42
wolfsprauland all names move in different ways, so a() becomes A() while at the same time B() becomes b()23:42
wolfspraulgo figure23:42
wpwrakwell, there's the big redesign of the internal database engine ...23:43
wolfspraulI don't mind cleanup, at all. but when I see names changing in different (!) directions, that's bad23:43
wpwrakof course, one problem may be that the task is too big to do it in one session. that's the day job problem23:43
wolfspraulso those big changes are happening in the live code base? or is there a branch somewhere?23:44
wolfspraulif it's in the one trunk codebase, maybe it can be finished one day :-)23:44
wpwrakand yes, redundant renames can be a bad sign. i do that quite a lot too, though. but it's usually quick. maybe 5 minutes and then i like my code again.23:45
wpwraki don't really know where those big changes are23:45
wpwrakmaybe ask ?23:45
wolfspraulI'm talking about inconsistent naming changes23:45
wolfspraulthe inconsistency is what is bad23:45
wolfspraulbecause you may as well stop the stupid renaming and first settle on *ONE* system23:45
wolfspraulotherwise I rename your names, and you rename mine23:46
wpwraki've seen some blueprints mentioned. and a bit of infrastructure code is getting merged. but i don't know if there's a real migration branch somewhere23:46
wolfsprauland we happily commit ever thereafter23:46
wpwrakah yes, if people are diverging, that's a problem23:46
--- Thu Mar 29 201200:00

Generated by irclog2html.py 2.9.2 by Marius Gedminas - find it at mg.pov.lt!